Merge pull request #269 from kerberos-io/fix/media-looping-boundary

fix/media-looping-boundary

machinery/src/video/mp4.go

@@ -56,6 +56,7 @@ type MP4 struct {
 	FreeBoxSize             int64
 	FragmentStartRawPTS     uint64            // Raw PTS for timing when to flush fragments
 	FragmentStartDTS        uint64            // Accumulated VideoTotalDuration at fragment start (matches tfdt)
+	LastKeyframeRawPTS      uint64            // Raw PTS of the most recently seen keyframe (in any fragment)
 	MoofBoxes               int64             // Number of moof boxes in the file
 	MoofBoxSizes            []int64           // Sizes of each moof box
 	SegmentDurations        []uint64          // Duration of each segment in timescale units
@@ -233,6 +234,24 @@ func (mp4 *MP4) flushPendingVideoSample(nextPTS uint64) bool {
 	var duration uint64
 	if nextPTS > 0 && nextPTS > mp4.VideoFullSample.DecodeTime {
 		duration = nextPTS - mp4.VideoFullSample.DecodeTime
+		// Guard against forward PTS jumps (e.g. when looping a source MP4
+		// through virtual-rtsp the upstream ffmpeg may insert a large offset
+		// at the loop boundary, or the RTSP stream may stall briefly).
+		// Without this clamp the sample gets a huge duration which appears
+		// as a discontinuity in the trun/sidx/mvhd and causes browsers
+		// (Video.js / MSE) to abort playback with a "media corruption"
+		// error around the loop boundary.
+		var maxPlausible uint64 = 1000 // 1 second hard ceiling
+		if mp4.LastVideoSampleDTS > 0 && mp4.LastVideoSampleDTS*10 < maxPlausible {
+			maxPlausible = mp4.LastVideoSampleDTS * 10
+		}
+		if duration > maxPlausible {
+			log.Log.Warning(fmt.Sprintf("mp4.flushPendingVideoSample(): video PTS jumped forward (nextPTS=%d, prevDTS=%d, gap=%d ms) - clamping to %d ms (likely source loop/stall discontinuity)", nextPTS, mp4.VideoFullSample.DecodeTime, duration, maxPlausible))
+			duration = mp4.LastVideoSampleDTS
+			if duration == 0 {
+				duration = 33
+			}
+		}
 	} else {
 		// No valid nextPTS (Close case) or PTS went backwards (jitter/discontinuity)
 		if nextPTS > 0 {
@@ -289,6 +308,27 @@ func (mp4 *MP4) AddSampleToTrack(trackID uint32, isKeyframe bool, data []byte, p
 		}
 		shouldFlush := !mp4.Start || elapsed >= FragmentDurationMs
 
+		// Detect upstream source-loop / restart discontinuity. When an MP4 is
+		// looped through virtual-rtsp (ffmpeg `-stream_loop -1 -re`) the loop
+		// seam emits a fresh IDR much sooner than a normal GOP would. PTS keeps
+		// growing monotonically, so the timing-only `elapsed` check above does
+		// not catch it and the seam IDR ends up as a mid-fragment sync sample.
+		// MSE-based players (Video.js / Chromium / Firefox) reject the resulting
+		// fragment with a "media corruption" error because the inner IDR resets
+		// frame_num/POC inside what they expect to be a single GOP. Force a
+		// fragment boundary whenever two consecutive keyframes arrive much
+		// closer than a normal GOP (here: < 500 ms apart). This isolates the
+		// seam IDR into its own fragment so each fragment stays a clean GOP.
+		if !shouldFlush && trackID == uint32(mp4.VideoTrack) && mp4.Start &&
+			mp4.LastKeyframeRawPTS > 0 && pts > mp4.LastKeyframeRawPTS &&
+			pts-mp4.LastKeyframeRawPTS < 500 {
+			log.Log.Warning(fmt.Sprintf("mp4.AddSampleToTrack(): forcing fragment flush at unexpectedly close keyframe (gap=%d ms, fragment elapsed=%d ms) - likely upstream loop/restart discontinuity", pts-mp4.LastKeyframeRawPTS, elapsed))
+			shouldFlush = true
+		}
+		if trackID == uint32(mp4.VideoTrack) {
+			mp4.LastKeyframeRawPTS = pts
+		}
+
 		if shouldFlush {
 			// Write the previous segment to the file
 			if mp4.Start {
@@ -393,6 +433,14 @@ func (mp4 *MP4) AddSampleToTrack(trackID uint32, isKeyframe bool, data []byte, p
 					if started {
 						dts = 1
 					}
+					// Guard against forward PTS jumps (e.g. virtual-rtsp loop
+					// boundary or upstream stalls). Without this clamp the
+					// audio trun would carry an enormous sample duration that
+					// renders the recording unplayable in browsers.
+					if mp4.LastAudioSampleDTS > 0 && dts > mp4.LastAudioSampleDTS*10 {
+						log.Log.Warning(fmt.Sprintf("mp4.AddSampleToTrack(): audio PTS jumped forward (pts=%d, prevDTS=%d, gap=%d) - clamping to last known duration", pts, mp4.AudioFullSample.DecodeTime, dts))
+						dts = mp4.LastAudioSampleDTS
+					}
 					mp4.LastAudioSampleDTS = dts
 					//fmt.Printf("Adding sample to track %d, PTS: %d, Duration: %d, size: %d\n", trackID, pts, dts, len(aac[7:]))
 					mp4.AudioTotalDuration += dts